Extent of Hyperbaric Spinal Anesthesia Influences the Duration of Spinal Block

After approval by our institutional ethics committee, informed consent was obtained from 40 patients classified as American Society of Anesthesiologists physical status I or II scheduled for orthopedic lower limb surgery with a thigh tourniquet during spinal anesthesia.

All patients received 7.5 mg of midazolam per os,  1 h before surgery. Monitoring equipment consisted of a pulse oximeter, a standard electrocardiogram, and a noninvasive arterial blood pressure device (AS3; Datex Cie, Helsinki, Finland). Ten milliliters per kilogram of lactated Ringer’s solution was infused 20–30 min before the beginning of spinal anesthesia, and infusion was continued at 5 ml/kg during the first 30 min after spinal injection.

Spinal anesthesia was performed in the sitting position using the midline approach at the L2–L3 or L3–L4 intervertebral space, with a 27-gauge Whitacre needle. After a free reflux of cerebrospinal fluid was obtained, 3 ml hyperbaric bupivacaine, 0.5% (Carbostésine 0.5% Hyperbare; Astra Pharmaceutica, Dietikon, Switzerland), was injected in approximately 15 s, with the bevel of the needle orientated in the cephalad direction. Using sealed envelopes, patients were allocated randomly to placement, immediately after spinal injection, in either a horizontal position or with the torso elevated 30°. The latter position was obtained by inclining the operating table at the level of the hip by 30°. After surgery, this position was maintained during transfer, as well as in the recovery room, by inclining the head of the bed by 30°, until the end of the study, defined as regression of sensory level below L4.

In the operating room an anesthesiologist and in the recovery room a nurse assessed the following variables:

• • Mean arterial pressure and heart rate every 5 min during surgery, then every 15 min until the end of the study.

• • Cephalad sensory level by loss of temperature discrimination with ether drops, every 5 min until 30 min after spinal injection, then every 15 min until regression below L4.

• • Motor block every 5 min until 30 min after spinal injection, then every 15 min until complete motor recovery, on a previously described modified Bromage scale 7as follows: 0 = able to move hip, knee, ankle, and toes; 1 = unable to move hip, able to move knee, ankle, and toes; 2 = unable to move hip and knee, able to move ankle and toes; 3 = unable to move hip, knee, and ankle, able to move toes; and 4 = unable to move hip, knee, ankle, or toes.

A 30% decrease in systolic blood pressure below baseline was treated with 5 to 10 mg intravenous ephedrine, and decrease of the heart rate less than 45 beats/min with 0.5 mg intravenous atropine.

From the recorded variables, we assessed the following time intervals, defined as time elapsed from spinal injection to:

• 1. Regression of sensory level by two dermatomes

• 2. Regression of sensory level to L2

• 3. Regression of sensory level to L4

• 4. Motor recovery by one level

• 5. Complete motor recovery (modified Bromage scale = 0)

• 6. Time to appearance of pain on the operative site

Postspinal headache and neurologic complications were recorded, until discharge from hospital.

Prospective power analysis indicated that 7–16 patients per group would allow detection of a 30% difference in duration of anesthetic block between the two groups, assuming a 20–30% SD of the duration of spinal block in the standard population. Continuous data are expressed as the mean ± SD and were compared using the unpaired Student t  test. Discrete data (such as sensory levels) are expressed as the median with range and were compared using the Mann–Whitney U test. Chi-square analysis was used to compare ordinal data. P < 0.05 is considered significant.

Demographic characteristics, American Society of Anesthesiologists status, and type and duration of surgery and duration of tourniquet inflation were comparable in both groups (table 1).

The main data on spinal block are summarized in table 2. Figure 1shows the evolution of median sensory level over time in both groups. The maximum cephalad spread of sensory block was significantly higher (P < 0.05) in the horizontal group (T3.5, range T1–T9) than in the group with torso elevation of 30° (T10, range T6–L1). The regression times by two segments and to segment L4, as well as time to complete motor block recovery, were significantly shorter in the horizontal group compared with the group with torso elevation of 30° (P < 0.05). In the elevated torso group, three patients did not have complete motor blockade; the anesthetic level was too low to block hip flexion efficiently, but operating conditions were excellent at extremities. These three patients were not considered in the evaluation of the motor blockade recovery. After regression of sensory level below L4, three patients in the horizontal group and seven in the elevated torso group did not present any pain at the operative site. Pain appeared significantly faster in the remaining 17 patients of the horizontal group than in the 13 patients in the elevated torso group (P < 0.05).

Figure 2represents evolution of mean arterial pressure and heart rate in the first 30 min of the study. The baseline mean arterial pressure and heart rate were comparable in both groups. Compared with the elevated torso group, maximal decrease in mean arterial pressure was significantly higher in the horizontal group after spinal injection (P < 0.05;table 3), without any modification of heart rate in either group. Four patients in the horizontal group required ephedrine. Atropine was never given, and neither anesthetic failure nor complications related to spinal anesthesia were observed.

This study was designed specifically to evaluate the relationship between the spread of a hyperbaric spinal anesthesia and its duration. Two groups with the same demographic and anesthetic characteristics (same dose, volume, and density of local anesthetic) were compared but put in two distinct positions to obtain significantly different spreads of spinal block (T3.5 vs.  T10), which validates our model. Our results confirm the hypothesis that the higher the spread, the shorter the duration of spinal block observed.

The influence of spread of a local anesthetic solution on duration of spinal block has never been studied specifically for hyperbaric solutions but has been suggested strongly with isobaric solutions. The specific aim of a study by Urmey et al.  8was to evaluate the effect of the direction of a Whitacre needle aperture during spinal injection on the spread of the injected solution. Sixty milligrams of isobaric lidocaine, 2%, injected in the cephalad direction resulted in a significantly higher anesthetic spread (T3) compared with the same solution and the same dose injected in the caudal direction (T7). More relevant, the duration of the spinal block was significantly shorter in the group with a higher cephalic spread compared with the group with a lower one (149 ± 31 vs.  178 ± 24 min), which is in agreement with our results. The authors do not comment on this finding, however, in their discussion.

The previously mentioned studies 2–6as well as the study of Urmey et al.  8suggest, and the current study confirms, that the spread of spinal anesthesia influences the duration of spinal block. These findings could be explained by the pharmacokinetics of local anesthetics injected into the subarachnoid space. As noted by Greene, 1local anesthetics are eliminated by diffusion across the arachnoidal and dural membranes from the subarachnoid to the epidural space, the concentration gradient being maintained by subsequent vascular absorption in the epidural vessels. Simultaneously, vascular absorption takes place in the subarachnoid space and in the substance of the spinal cord. The amount of bupivacaine available to block a spinal segment is low if cephalad spread is high, compared with a restricted spread, with a higher amount of agent per segment. With a greater extent of anesthesia, elimination of local anesthetics from the sites of action should be enhanced by a greater meningeal surface, although balanced by a lower gradient of concentration. Burm et al. 9demonstrated by a marked isotopic method that local anesthetics were not metabolized in the subarachnoid space, because the diffusion and the vascular absorption of hyperbaric lidocaine and bupivacaine from the subarachnoid space to general circulation was complete.

Sacral pooling of a hyperbaric local anesthetic solution could be enhanced by a 30-degree elevation of the torso. A higher localized concentration of local anesthetic has been shown to induce neurotoxicity, as classically described with lidocaine spinal anesthesia. 10In the present study the regression of the sensory block to L2 and the recovery of motor block by one degree, though faster in the horizontal group, were not statistically different from in the elevated torso group. Although the difference becomes obvious if the regression to L4 or the total motor recovery are compared, we can only speculate about the risk of enhanced sacral accumulation of hyperbaric bupivacaine. No residual neurologic symptoms were noted in this study. Furthermore, transient radicular irritation rarely is described with 0.5% hyperbaric bupivacaine. 11,12 

The decrease in the mean arterial pressure was significantly more severe in the horizontal group. This finding is in agreement with previous studies 3,6and can be related to the extended sympathetic block following the higher cephalic spread obtained in this group.

For methodological reasons, the present study was not realized in a double-blind fashion, but the persons assigned to test the patients were unaware of the aim of our study. The sensory block was assessed by disappearance of temperature discrimination using ether drops, a test currently used in our institution to determine the spread of sensory blockade with different regional anesthetic techniques. Compared with the pinprick technique, we consider the temperature discrimination test more accurate, less aggressive, easier to perform, and easier to reproduce if performed by different observers. Loss of temperature discrimination occurs one or two dermatomes higher, but its onset and regression closely parallel those of the sensory level assessed by pinprick. 13 

In conclusion, our results confirm the hypothesis according to which the duration of spinal blockade with the same dose of hyperbaric bupivacaine is longer if the spread of the local anesthetic is restricted. Thus, for hyperbaric solutions, the position of the patient influences not only the spread, but also the duration of the spinal blockade. Consequently, placing the patient in a position with 30-degree elevation of the torso after spinal injection prolongs the duration of both sensory and motor blocks. On the other hand, placing the patient in a horizontal position allows a shorter duration of spinal block. However, the hemodynamic consequences of an extensive spread in the horizontal position should be weighed against the risk of sacral pooling with the 30-degree torso elevation, which may lead to neurotoxicity.